[IBIS-Users] RE: [IBIS] SI_location / Timing_Location

From: Muranyi, Arpad <Arpad_Muranyi@mentor.com>
Date: Tue Aug 27 2013 - 10:48:44 PDT
Thanks for all who responded so far.

I got the general idea, but what I am still missing
is a list of parameters or subparameters in [Model]
or [Model Spec] (or any other keyword) on which these
two parameters "operate".  Do we have an unambiguous
answer to that by reading the IBIS specification?

Thanks,

Arpad
======================================================

From: Scott McMorrow [mailto:scott@teraspeed.com]
Sent: Tuesday, August 27, 2013 12:39 PM
To: Robert Haller
Cc: Aubrey Sparkman; Muranyi, Arpad; IBIS; IBIS-users
Subject: Re: [IBIS] SI_location / Timing_Location

Actually, Si_location and Timing_location go back to 1997 and were introduced by Bob Ross of Interconnectix (pre-Mentor)

http://www.eda.org/ibis/birds/bird43.txt

Timing_location may also need to be changed to die for large FPGA type devices, because of large non-monotonic signals at the pins.



On Tue, Aug 27, 2013 at 1:30 PM, Robert Haller <rhaller@enterasys.com<mailto:rhaller@enterasys.com>> wrote:
I think the SiSoft guys probably pushed for this a while ago. Their
software can do chip to chip timing either at pin or pad (on either end of
transmission line). Usually one performs static timing from pin to pin
because then there is no double counting when using data sheet timing
specs. But if you are designing the silicon and the package, there are
good reasons to perform timing pad to pad when you are trying to shave
picoseconds.
Regards
Bob

-----Original Message-----
From: owner-ibis@eda.org<mailto:owner-ibis@eda.org> [mailto:owner-ibis@eda.org<mailto:owner-ibis@eda.org>] On Behalf Of Aubrey
Sparkman
Sent: Monday, August 26, 2013 2:19 PM
To: Muranyi, Arpad
Cc: IBIS; IBIS-users
Subject: Re: [IBIS] SI_location / Timing_Location

I have used these switches, though I'm not sure I gave them different
values.

Possible scenario...   Interface specs timing at pin, but I want to make
sure edges are monotonic at the silicon.

Just a though.

Aubrey Sparkman

Sent from my iPhone

On Aug 26, 2013, at 11:44 AM, "Muranyi, Arpad" <Arpad_Muranyi@mentor.com<mailto:Arpad_Muranyi@mentor.com>>
wrote:

> Hello IBIS experts,
>
> Someone asked me the question whether Si_location and Timing_Location
> can have two different values because they have an IBIS file which
> does that.
>
> In attempt to answer their question, I searched the IBIS specification
> for Si_location and only found this sentence on pg. 16 (in v5.1):
>
> "Si_location and Timing_location are optional and specify where the
Signal Integrity and Timing measurements are made for the component.
Allowed values for either subparameter are "Die" or "Pin". The default
location is at the "Pin"."
>
> I found no other occurrences of "Si_location" or "Timing_location"
> in the entire specification, i.e. their meaning, usage and purpose are
> not defined.  Or am I missing something?
>
> Going back to the original question, does it make sense to have two
> different values for these two parameters?
>
> Thanks,
>
> Arpad
> =====================================================================


-- 
This message has been scanned for viruses and
dangerous content by MailScanner, and is
believed to be clean.


--------------------------------------------------------------------
|For help or to subscribe/unsubscribe, e-mail mikelabonte@eda-stds.org
|or ibis-request@eda-stds.org
|
|IBIS reflector archives exist under:
|
|  http://www.eda-stds.org/ibis/email_archive/ Recent
|  http://www.eda-stds.org/ibis/users_archive/ Recent
|  http://www.eda-stds.org/ibis/email/         E-mail since 1993
Received on Tue Aug 27 10:49:11 2013

This archive was generated by hypermail 2.1.8 : Tue Aug 27 2013 - 10:49:25 PDT